Bret Weinstein and a web of lies
Friends just keep on sending me Bret Weinstein videos.
To summarize his argument, Bret thinks that ivermectin cures covid if you’re sick and is nearly 100% effective at preventing covid¹. He thinks that big pharmaceutical companies are suppressing these facts to profit from more expensive drugs and vaccines². Government organizations like the FDA and WHO are all compromised and controlled by big pharma³. Facebook and Google are in on the plot, censoring any information in favor of ivermectin. The covid vaccines are more dangerous than anyone wants us to think⁴. And we don’t need the vaccines anyways, because if we all took ivermectin, the virus would go extinct⁵.
You can find most of those claims in the this podcast he did with vaccine skeptic Steve Kirsch. It’s a lot to unpack. Some arguments in favor of the drug are plausible, others are really bad and easy to debunk. Let’s start with the bad ones.
Ivermectin’s biggest proponent is Dr Pierre Kory. Kory generally seems credible, with medical training, experience treating covid, and a history of advocating for a drug which was later approved for covid (dexamethasone).
Dr Kory describes Ivermectin as a miracle drug. He thinks it’s 50% effective against covid in late treatment, 75% if given early, and 80% effective at preventing covid entirely, if you take the drug preemptively.
Kory goes further to claim that other countries have defeated covid with ivermectin use. He says that Mexico’s hospitals emptied out because they used the drug. He says that India beat covid with ivermectin.
The first thing to notice is that Mexico’s winter covid spike was about the same as the United States:
The whole northern hemisphere dealt with covid last winter. How do we know that Mexico’s death numbers went down because of ivermectin, when other countries had the same trajectory? The covid wave might have ended for lots of other reasons: lockdowns, distancing, masks, seasonality of the virus, herd immunity.
Kory’s organization, the FLCCC, compares states in India to “prove that ivermectin works”. One state (Tamil Nadu) did not use ivermectin, many other states did. The FLCCC shared this graph, claiming that every state except for Tamil Nadu beat the virus:
Can you guess what happened next? Did cases in Tamil Nadu blow up forever, or did they spike and then decline?
Cases in Tamil Nadu peaked a week after where the FLCCC graph ends, without using ivermectin. Their graph was cropped to not show the decline.
Bret Weinstein shared the FLCCC’s graph on June 8th, 2 weeks after Tamil Nadu’s cases were well into the decline. Either he didn’t check the data for himself or he didn’t care.
A number of other states in India peaked around the same time as Tamil Nadu. Andra Pradesh peaked late. So did Manipur, which was using ivermectin.
Goa has since stopped prescribing ivermectin but cases haven’t gone back up.
Relative to population size, Tamil Nadu’s spike was smaller than other states in India, and Goa had the biggest spike:
So, the FLCCC is cherry picking data and cropping their graphs to try to make a bad point. Every state in India had a case spike and then a decline, regardless of ivermectin use. Peaks happened at different times, I don’t know why. It could be different weather, different virus strain, different lockdown rules. We can graph social distancing across different Indian states:
That data makes it look like Tamil Nadu just locked down later than other states. That alone doesn’t disprove the ivermectin theory, but it does give us an alternate theory: every state locked down when covid hit, then covid cases went down because of lockdowns, not because of ivermectin. A third theory would be that neither ivermectin nor lockdowns did much of anything: the virus just came through, infected everyone, and then the case spike ended.
There’s no reason to think that every state in India should have exactly the same curve. Every state in the US has a different curve, and people intuitively understand that. When the FLCCC starts talking about India, people are mystified and believe it’s all about ivermectin.
As I understand it, ivermectin is being used for treatment, not for prevention. The main signal we should be looking at isn’t in the number of cases. It’s the death rate of the disease.
Ivermectin proponents say the drug is 75% effective. So, all else equal, there should be less deaths per diagnosed covid case. The case fatality rate should go down.
We can graph the CFR, over time, and compare countries:
The CFR for the US fluctuates between 1 and 2% with no obvious pattern. CFR depends on the true death rate and the rate of testing, so this is a noisy signal. India’s rate is similarly noisy, but there’s no decline in fatality from ivermectin use this spring.
One ivermectin proponent on Twitter claims that Panama fought off covid with the drug. The country started using ivermectin for treatment in mid January. He graphs South America as the control group:
Panama’s CFR is also just random noise between 1 and 2%:
Mexico city started treating patients with ivermectin on December 29th. Anyone with a positive test would be sent home with a few ivermectin pills (each patient got 4 pills, enough to last 2 days, plus some tylenol and some aspirin).
We can graph the case fatality ratio for Mexico City and the rest of the country:
There might actually be a signal here. There’s no big change in CFR for December or January, definitely no quick 75% reduction in mortality. But there is a 50% reduction in CFR in Mexico from April to June. It’s of the same size as the random drawdowns in the US graph but looks more sustained.
Mexico city has seen an even larger drawdown since April, maybe on the order of 75%.
If Mexico’s progress came from ivermectin, the drug wasn’t rolled out widely until April. Given the timing, it’s important to also post this graph:
Elderly people start getting vaccinated at the end of February, they get immunity by the end of March, CFR starts declining early April because the remaining cases are younger and healthier. Based on the timing, this makes more sense than the ivermectin explanation.
Panama and India show no clear evidence that they’re improving care. Either ivermectin isn’t working, or they’re not using it widely enough.
One thing to make clear is that none of these countries have beaten the pandemic with cheap drugs. The clearest signal of the pandemic is excess deaths, compared to an average year. The US is doing better than Mexico or Peru, by this metric. Covid is bad in most countries, it’s worse in Latin America.
I can’t put India on the same graph, because no one in India even keeps track of how many people die in a year.
Researchers have tried counting deaths in regions of India and extrapolating. One reasonable guess is that 2 million Indians have died from covid.
Say what you want about your own government’s covid response. At least they’re not failing to report over a million deaths.
Edit (11/12): I went back and checked some more recent data. Dr Kory claimed that Mexico had emptied out their hospitals by using ivermectin. Covid death rates went right back up when delta hit Mexico. Covid deaths in Mexico continued to be about the same as the US:
Dr Kory also claimed that Zimbabwe had eradicated covid with ivermectin:
On July 2nd, Kory said that Slovakia had beaten covid with ivermectin, and said “The world could literally end Covid tomorrow by following their lead.”
A few months later, deaths in Slovakia started going up again:
Are you seeing the picture yet? Find some obscure country where covid case numbers are going down. Tweet that ivermectin is the cause. Most people will never go back and check.
So, the population level data is mostly crap. Every country that’s seen as an ivermectin “success story” is doing worse than the US, overall. These arguments are designed for a very gullible audience.
Drug trials tell a mixed story. Last year, scientists found that ivermectin kills covid-19 in vitro. That is, if you grow cells in a petri dish and infect them with covid, ivermectin can stop the virus from replicating. But there’s some controversy over whether taking a few ivermectin pills gets cells up to a high enough dosage to have this effect. This is confusing because it depends which type of cells you look at. The original study used kidney cells from a monkey, and just looking at those results, it’s unlikely that ivermectin should work at all. A more recent study on human lung cells showed different results and a chance that the drug might work. The math is complicated so I’ve left it as a footnote⁶.
Basically, the in vitro evidence is inconclusive. With the least favorable studies, the drug doesn’t work at all. With the most favorable assumptions, a large daily dose could work for treatment. Even with the most favorable assumptions, it seems unlikely to me that a small dose would prevent covid.
To really understand it, we need to test it on people. Human trial results have been mixed but they lean positive. In the largest randomized experiment, 200 ivermectin treated patients recovered a little faster than the control group, in 10 days instead of 12. Not a miracle cure, but it would still be nice if that’s true. The results were not statistically significant, that could also just be a random result.
One popular website, Ivmmeta.com, gives a summary of all the trials. At first glance, I have to say that this website reminds me of an election fraud website called hereistheevidence.com:
When people argued about fraud, after the 2020 election, liberals would say there’s no evidence. Conservatives would point to a website called “here is the evidence”. It included dozens of fraud claims, anecdotes, and abnormalities. Many of the claims were easy to disprove, with a bit of digging, but the website conveyed a sense that there was a huge amount of data and few people had the patience to go through it all.
An election fraud lawsuit claimed there was a “one in a quadrillion chance” that Biden had won, based on all the evidence. The ivermectin website says:
“The probability that an ineffective treatment generated results as positive as the 63 studies to date is estimated to be 1 in 1 trillion.”
So, the language used seems suspicious, but it’s still worth looking at their data. There’s a lot going on in this table, so I just want you to focus on a few things. First, notice that almost every study had positive results (green square is left of center). Second, notice that the number of patients in the treatment and control groups are small, typically less than 100 patients, so the error range around each green square is large. The error range usually includes the possibility that the drug doesn’t work. Third, notice that for each trial, they pick a different measure of success, whether that’s hospitalizations or time to recovery or PCR testing.
They could be picking different measures of success to make the data look stronger than it is. There could also be publication bias here. Small studies return random results, maybe they only get published if the results are positive. But the evidence still seems mostly positive, and it seems to tells a consistent story. Early treatment is better than late, and prevention is better than early treatment.
If you think that meta-analysis is ironclad proof, I’d encourage you to look at another meta-analysis which shows that aspirin cuts covid deaths by 50%. Either, meta-analysis is easily fooled. Or, there are lots of early treatments that might help with covid and we’re doing poorly at using any of them. Aspirin could work because covid causes blood clots and aspirin reduces clotting risk. Other non-prescription treatments that might help include vitamin C, vitamin D, zinc, famotidine, melatonin, quercetin, and bromhexine/ambroxol (not available in every country).
Ivermectin trials gave better results for covid prevention than for treatment. Reading the prevention trials, I can see why some scientists are skeptical. Behera’s large trial uses self reporting of symptoms to diagnose covid, not covid testing. Carvallo’s trial, where prevention is 100% effective, tests ivermectin and another drug (iota-carrageenan nasal spray) at the same time without trying to figure out which one is working. The trial subjects (all health care workers) use the experimental nose spray 5 times a day, so they’re going to be doing different things with regards to their hygiene and masks, compared to the control group. (Edit: More recent evidence suggests that this trial never even took place, the claims that ivermectin is 100% effective might be totally fabricated)
Good science shouldn’t be this hard. Pick one variable. Change that variable and nothing else. Randomize who gets the drug. Get enough patients that the results are significant. I don’t see one trial like that in the whole set.
The strongest evidence I’ve seen comes from a very large study done in Mexico City. 80,000 patients were sent home with tylenol, aspirin, and four ivermectin pills when they first got their covid test. They were 75% less likely to end up in the hospital than similar patients who were not treated. The study wasn’t randomized, the researchers just tried to match patients who were sent home with drugs in January against case records from a month earlier, when patients weren’t treated. The study is a preprint, they don’t provide much source data, it could be confounded in some way. It’s possible that the aspirin is doing some of the work.
A 75% reduction in hospitalization would be huge and important. Doctors should take notice. They should try the drug more, they should test it faster, they should see if they can reproduce this success elsewhere. A million lives could be saved, over time. It would only take one large randomized controlled trial to prove this, one way or another. Larger trials are underway.
For now, the important questions are how we deal with that uncertainty. Should we prescribe a drug if it’s unproven, but there’s a chance that it helps? And should we be allowed to talk about it?
Weinstein is being censored by Google and Youtube. They do it in blatant ways, like taking down 2 of his videos and taking away all ad revenue from his Youtube channel. They do it in subtler ways as well. Search Google for a recent episode of his podcast and you’ll instead see an obscure comic book from the 90’s as your first hit:
Google asks: did you mean “darkhorse” like you searched for? So you can click “search instead for darkhorse 84”, without the space in the middle. And the comic book is still the first hit:
And that’s fishy, because on DuckDuckGo, the comic book doesn’t even show up on the first page, either way you spell it. It’s almost all Bret’s podcast:
So, why is Weinstein being censored by Google and Youtube? Is it all a conspiracy?
Weinstein’s theory is that big pharma, government, and tech are working together to suppress this drug. And they’re doing it to focus on more profitable drugs like Remdesivir. Ivermectin costs pennies a pill, while Remdesivir costs $3000 for a course of treatment. I can see some logic there, Gilead has an incentive to oppose the generic drug.
The logic breaks down, the more you look. The World Health Organization, part of Weinstein’s conspiracy, now recommends against using Remdesivir for covid. The government isn’t widely promoting expensive antibody treatments, like the $1,000 Regeneron drug that Trump took when he got sick, they’re using cheaper vaccines. The AstraZeneca vaccine sells for only $4 a dose. The company is not going to get rich from that, distributing that to the whole world likely costs the same or less than giving everyone ivermectin. Pfizer’s vaccine sells for $20. Added up across a few billion doses, it’s a nice profit, but it won’t be their best drug ever.
Weinstein argues that the vaccines have killed thousands of people and the government is covering up all the deaths. But Europe and Australia both stopped using the AstraZeneca shot as soon as they discovered a 1 in a million chance of death from blood clots. I have trouble imagining a government that’s simultaneously that cautious and reckless.
Let me offer an alternate theory to Bret’s: ivermectin was ignored because everyone was arguing about hydroxychloroquine last year, instead.
There are lots of drugs that might help covid patients. The medical establishment is slow and cautious at testing all of them. One pro-ivermectin site gives a summary:
Ivermectin has been tried on 19 thousand patients in trials. Remdesivir on 30 thousand. But hydroxychloroquine has been tried on almost 400 thousand.
We can see the popularity of each in Google trends, over the last 18 months:
It wasn’t until December 2020 that ivermectin started getting more search traffic than hydroxychloroquine.
We know that the FDA will approve cheap, generic drugs for covid. Dexamethasone is a cheap generic drug that’s now used for covid treament. It was quickly approved after a trial proved that it worked.
Hydroxychloroquine was quickly given Emergency Use Authorization last March. But the EUA was revoked in June, after larger trials showed it did not work. From the Oxford RECOVERY trial:
A total of 1542 patients were randomised to hydroxychloroquine and compared with 3132 patients randomised to usual care alone. There was no significant difference in the primary endpoint of 28-day mortality (25.7% hydroxychloroquine vs. 23.5% usual care; hazard ratio 1.11 [95% confidence interval 0.98–1.26]; p=0.10). There was also no evidence of beneficial effects on hospital stay duration or other outcomes.
Hydroxychloroquine proponents said this was because they gave the drug too late.
Another study showed that hydroxychloroquine doesn’t work to prevent covid. You can’t take it any earlier than that.
Supporters kept moving the goal posts. It only works if you add zinc. Or azithromycin. Or the dose of hydroxychloroquine used in the studies was too high (like… it’s safe and it kills covid but you can’t use too much of it?)
They could be right. It could have some mild effect if used at just the right time. It could only work in some particular combination. But it’s not clear to me that it does anything at all.
Hydroxychloroquine got emergency approval. It was tested on 400,000 people. Doctors were obviously willing to entertain the idea that a cheap drug could work.
The problem is that it didn’t work.
If you ask people what they think about this, the responses vary, and they split along tribal lines. Many Republicans still say that the drug obviously works, and it was only suppressed to hurt Trump. Democrats say that the drug obviously failed, and Trump was an idiot for promoting it.
We have a culture war where each side is more concerned about winning points against the other team than they are about finding truth. And this culture war is harmful. Democrats need to be open to testing a cure, even if the idea comes from conservatives. Conservatives and contrarians need to be open to giving up on an idea once it’s disproven. If they’d walked away from hydroxychloroquine sooner, maybe we would have already moved on to testing more promising drugs.
So now, the debate has shifted to ivermectin, and people are again dividing into 2 opposing camps. We see a bunch of people saying that it obviously works and it’s being suppressed. We see a bunch of articles saying it’s just the next hydroxychloroquine.
Google has decided to take the left wing side on any culture war issue. And they’ve decided to create a list of things you can’t say about the pandemic. You can’t say covid is just the flu. You can’t say that masks don’t work. And you can’t say that ivermectin cures covid. Even though we have 60 small trials suggesting that it helps.
Google might save lives, by suppressing bad information. But, if they’re wrong about ivermectin, they’re helping kill 75% of covid patients.
Weinstein might save lives if he’s right. But he’s also scaring people away from taking the vaccine. There’s a body count associated with what he’s saying.
With all the debate, there’s room to make money off the controversy. Which raises the question: is Weinstein actually here to warn humanity that we’re missing out on a miraculous drug? Or is he just in this to get clicks, pitching controversial ideas to a contrarian audience to get millions of views? He says that he’s lost money, because Google demonetized his videos. But he also gained 2,000 people donating to his Patreon.
Controversy brings attention, and attention makes money. The attention doesn’t necessarily help the world. It’s not his audience that needs convincing. It’s doctors and scientists and decision makers. To convince those people, you need good science and good arguments, not lies about miraculous results in India. You need to conduct good trials and tell the truth about what you find.
The media may be ignoring ivermectin, but the medical establishment is not. There are several large randomized controlled trials in progress. Oxford’s PRINCIPLE trial is testing the drug. So is the University of Minnesota.
The medical establishment is slow. If the drug works, we won’t know for a few months. There are other drugs that are even more promising for covid treatment, like Fluvoxamine. At the rate things are going, we won’t have large trials of those ready for a year.
That speed of testing should be the point we’re outraged about. That’s the one thing we could actually fix.
So, we’ve got a drug that might work, but we don’t know for sure for months.
What do we do in the meanwhile? Is it riskier to use it? Or is it riskier to let people die from covid, without treatment?
The FDA is paranoid about side effects. European regulators are, as well. The AstraZeneca vaccine was pulled from use across Europe after regulators found it kills one person out of a million through blood clots. It’s still safer than many drugs we use on a daily basis, it’s much safer than getting covid, but the regulators stopped using it until they could print the right warning label. Australia is delaying its use until young people can get the Pfizer vaccine instead.
In the US, the AstraZeneca vaccine was never approved to begin with. The phase 3 trial data had some uncertainty, regulators decided it would be better to wait a few months until there were enough Pfizer doses. Thousands of people died during those months.
The FDA’s most famous success story came from Thalidomide. Thalidomide was a sedative used in Europe and Canada. It was marketed towards pregnant women as a cure for morning sickness. And it turned out to be toxic, causing malformed limbs in 1 out of 10,000 babies born. The FDA blocked approval for several years, and prevented harm to thousands of children in that time.
There are risks to using a new medicine, and there are risks to not using that medicine. At the moment, we’re living through a deadly pandemic, so the risks of not using medicines are much higher.
Some people have died from covid vaccination. The numbers could be in the hundreds, they’re probably not in the thousands. For every day we delayed approving covid vaccines, a few thousand people died from the virus.
Every death from vaccine side effects added up would be less than a single day of covid deaths.
If you gave ivermectin to millions of patients, you might kill someone with side effects, or unknown drug interactions with every other pill they’re taking. It would still be safer than the virus.
So, I’m pro-vaccine because the reward is much greater than the risks. And I’m also in favor of trying promising drugs more.
This is a nuanced position. It depends what the alternatives are. If you’re a doctor in India, you have thousands of sick patients, and there aren’t enough vaccines to go around, it makes sense to just try ivermectin or other unproven treatments. If you’re a person in the United States, where we have enough vaccines for everyone, and you’re choosing to take ivermectin instead of the vaccine, that doesn’t make much sense.
The main thing we need to do is collect data when we try these treatments. We could make it legal for your doctor to send you home with ivermectin or other promising drugs. You heard about it on Youtube, you think it works, your doctor isn’t sure. You sign a form waiving any liability if it harms you, your doctor sends you home with some pills. We track these patients against people that weren’t treated. If it’s 75% effective, we’ll quickly see that in the data. And if it doesn’t work, it’s still better than you going and buying veterinary ivermectin and taking it with zero supervision.
We could also just do more randomized trials, but streamline the process to test each drug in 1–2 months.
Big pharma has no incentive to fund trials for generic drugs, so the government needs to step in and fill the funding role.
Instead of fighting a six month culture war over every new drug, we’d be better off coming together to build a faster system. Republicans would be happy because they might get a pill that works and they won’t have to take a scary new vaccine. Democrats would be happy because this system would spend more tax money.
We need both covid treatments and vaccines.
Many countries won’t be able to vaccinate people for a year or more. They need any covid treatments they can get, until that happens. A 75% effective treatment would save over a million lives.
We will also get some breakthrough cases of covid in vaccinated people. Those people need treatment.
And we might still get a covid variant that bypasses the current vaccine. Then we’ll need something, while we scramble to revaccinate everyone.
If ivermectin does work, and it works more universally against covid variants, then the drug would be very useful. If it doesn’t, we need to move on to testing the next option, whether generic or not.
Either way, ivermectin will not be a panacea, and it will not replace vaccines. Bret Weinstein exaggerates here and claims that we can rid the world of covid if we get everyone taking the drug.
It should be clear that this is nonsense. Covid spread from a single person in China to the entire world in under 6 months. To eradicate it, we’d need to get to zero cases, world-wide.
We would need widespread compliance. We can’t even get half of Americans to sign up for 2 vaccine shots. Do you really think we can get them all to take pills weekly? And give those pills to their kids? And maybe also use a nasal spray 5 times a day?
Some people would be too lazy to get the prescription, others not diligent enough to take it regularly. Some would make memes that the pills are toxic and part of a Bill Gates plot to sterilize us. Facebook would ban the memes. And people would see that as evidence that they were right and Facebook is in on the plot.
Beyond that, we’d have to get everyone on the planet taking it. Even in underdeveloped and unstable areas that can’t eradicate diseases like polio.
We’d have to eradicate it from animal populations, since the disease can spread in some animals, like Mink. And we’d need to destroy every laboratory sample, or develop perfect safety, in every lab in the world.
If we leave a single person or animal infected, it could spread to the entire world again. And we’d all be back to taking pills.
There is no magic bullet. Covid isn’t over until we all have antibodies. And it might still not be over then. It depends how long your antibodies last. It depends how quickly it evolves to evade the immune system. We might have to revaccinate yearly for new mutations.
My advice? Just get your covid shot and enjoy being immune for now. Better trials will be in soon, and we’ll know by then if ivermectin works or not. Until then, quit sending me Bret Weinstein videos.
Update (November 12th, 2021):
A lot has happened in the months since I wrote this. Let me try to summarize the few biggest changes.
The first big randomized controlled trial for Ivermectin came in and the results were unimpressive.
There was a 10% reduction in hospitalizations and that result might have been explained entirely by chance:
These trial results were met on Twitter with huge amounts of cope. People used every excuse you can imagine to explain the trial’s failure:
The BIRD group tweet is the best, because it implies that ivermectin is so good that it still works a month after taking it. Twitter later banned the BIRD group.
The Together trial, allegedly biased against cheap drugs, did show that cheap, generic fluvoxamine works well for covid, with somewhere between a 30–60% reduction in hospitalizations, depending on how you measure it. Some people didn’t finish taking the drug for 10 days, perhaps because of side effects. The people that finished their course of drugs did better than those who didn’t.
A number of older trials for ivermectin have been alleged to be fraudulent, including the trial that claimed it was 100% effective against hospitalization.
That should make us skeptical that the drug works.
On the pro-ivermectin side, I’ve seen two studies that made me think the drug could still work in certain doses. One in vitro experiment showed that ivermectin works in lung cells, in lower concentrations than we’d previously thought. I’ve discussed the significance of that results in Footnote 6.
There was also an experiment done in Argentina, where 3,000 patients were given high dose ivermectin. Those patients were 50% less likely to get hospitalized than the “control group”. The experiment was not randomized, so there isn’t really a control group, they just tried to pick out a group of similarly aged patients who weren’t treated with the drug. The group they chose isn’t necessarily well matched, there are always confounders. The few confounders I looked at don’t seem to explain away the result — like, the patients taking ivermectin were fatter than the patients who weren’t, so that strengthens the result, it doesn’t explain it away.
That experiment is large enough that the result they got can’t be chance. It could be confounded in some way. More than anything, I’d call the experiment a colossal waste of time. If they’d simply randomized half of those 3,000 ivermectin patients to placebo, they would still have enough patients to prove without a doubt if the drug works or not. Even assigning only 1 out of 3 to placebo would have been good enough. But, the researchers didn’t do that, so every health authority in the world can dismiss the experiment as “not randomized”.
That study does raise the possibility that maybe the drug works, but only at high doses. The study in Argentina dosed patients differently than the Together trial done in Brazil. The former used 0.6 mg/kg for 5 days, and gave patients the drug with a meal. The latter used 0.4 mg/kg for 3 days and gave the drug on an empty stomach. Giving the drug with a meal should increase absorption, studies disagree on exactly how much.
So, where does that leave us?
There is still no strong evidence that ivermectin works. There is no RCT proving it. Many of the promising studies done in 2020 have been exposed as fraudulent and the only large RCT (the Together trial) showed the drug doesn’t work very well, maybe doesn’t work at all.
The drug did not save India and Africa and Mexico from covid. It’s not 100% effective at preventing covid. I’m not sure it prevents covid at all — the doses people use for prevention don’t seem high enough that they should work.
But the drug has some known antiviral properties and it’s still possible it can treat covid in some high dose regimes.
It’s possible we have a drug with some mild anti-viral effects and we also have a cult built around the same drug with wildly exaggerated beliefs about its effectiveness.
That cult is not going to give up on ivermectin. There are 3 more large, randomized trials coming up which should test the drug (Activ-6, Oxford, and COVID-OUT at University of Minnesota). Ivermectin supporters have already said they think all of these trials are designed for failure.
So we might not ever resolve this argument. If the trials all come back saying that ivermectin works well, we’ll all agree that it works and start using it. But if the trials come back with weak or null results, then ivermectin supporters will just call the trials rigged or dosed wrong and say the medical establishment is corrupt.
My guess is that all 4 big randomized trials will return weak or null results. Ivermectin supporters will call the results fake or rigged. And, like hydroxychloroquine, some supporters will never give up.
In the end, the only question I really have is, “how should I treat covid, if I or someone I know gets it?”
The FLCCC, an ivermectin advocacy group gives guidelines for treatment (see here, page 2).
I would personally suggest you read their guidelines in reverse.
Start at the end where they suggest maybe getting vaccinated, almost as an afterthought. I would do that, if you don’t already have immunity.
Since I first wrote this piece, we’ve learned that covid vaccines don’t last as long as we hoped. You can still catch covid after getting vaccinated. But your odds are lower and you’re likely to have a much weaker case of it. Your odds of getting hospitalized are still maybe 70–90% lower.
In the most charitable interpretation of the data, high dose ivermectin could be 50% effective against hospitalization. You get better protection than that just from being vaccinated.
If you do get sick and you’re worried about severe illness, your next best bet is getting monoclonal antibodies. One study shows those are 60% effective against hospitalization.
If you can’t get those, I’d consider asking your doctor about fluvoxamine, which was proven to work by the Together trial.
Now that you’ve chosen one of those 3 treatments that are actually proven to work, I’d move on to the middle section where they recommend aspirin, vitamins, and supplements. All of those things are easy to buy, they’re not going to hurt you and there’s some chance they could help you. I would take them.
At the beginning, they recommend Ivermectin and Nitazoxanide. Both drugs can fight viruses in a petri dish, neither has been proven to treat covid patients. I don’t see why those would be anyone’s first choice.
But… it’s your body and your choice. I’m just trying my best to interpret the science and debunk some of the worst misinformation.
Update (August 17th, 2022):
I was way too charitable with Bret Weinstein in this article. Bret has since gone full anti-vax grifter. Rather than apologizing for getting any of this stuff wrong he’s only admitted to changing his mind twice: he now thinks that hydroxychloroquine treats covid and that masks don’t work.
A few more big trials have come in for ivermectin.
Activ-6 published results. Ivermectin had zero effect at preventing hospitalization from covid. There might be a slight effect, people taking ivermectin recovered from covid in 11 days instead of 11.5. Or that might have just been statistical noise.
COVID-OUT also published results. Ivermectin failed to treat covid, at a dose of 0.4 mg/kg for 3 days. Half the patients got ivermectin in 3 days or less of getting sick, so this isn’t a problem with dosing too late. Ivermectin supporters could still argue that the dose was too low.
The COVID-OUT trial did find surprisingly good results for a drug called metformin, which cut hospitalization down by about 40%.
There’s something deliciously ironic about this. The RCT found a cheap generic drug that works for covid. It wasn’t ivermectin or hydroxychloroquine. It was a drug that almost no one talked about at all.
You could call metformin a dark horse.
In fact, the only article I ever remember reading about metformin used exactly those words. From July 31st, 2021, right in the middle of the culture wars about ivermectin:
Metformin: A Pharmaceutical Dark Horse in the Face of COVID-19
An old drug with new tricks?
medium.com
Bret Weinstein, host of the dark horse podcast, missed the drug because he was too busy selling his audience a treatment that was more popular.
I can’t think of a better way for this story to end.
Update (December 26th, 2022):
Ivermectin failed in an RCT at 0.6 mg/kg, for 6 days. That’s at least as much as the FLCCC wanted you to take. Patients recovered at the same speed as the control group, there was no improvement on hospitalization or death.
Dr Kory now recommends that children shouldn’t get any childhood vaccines, even for things like measles or tetanus. And Bret Weinstein is considering the idea that HIV doesn’t cause AIDS.
I think I’m done updating this article. At this point, I’m just beating a dead dark horse.
Footnotes:
1. At 3:40 in this podcast, Bret says ivermectin is 100% effective at preventing covid.
2. Bret has regularly discussed a theory that the covid vaccines are under Emergency Use Authorization and pharmaceutical companies are suppressing ivermectin to keep those EUA’s intact. At 56:57, he suggests that Merck (creator of Ivermectin) is suppressing the drug in order to profit from their new covid drug (Molnupiravir) as well as perhaps in collaboration with the Johnson and Johnson vaccine.
3. Bret refers to this process as regulatory capture. He claims that pharmaceutical interests control governments and other organizations. At 2:59:58 in the podcast, Bret and his guests claim that pharma has power over journals, universities, social media, organizations of doctors, and the WHO.
4. At 1:01:38, Bret and Steve Kirsch discuss vaccine deaths, saying there are between 5,000 and 500,000 in the US. We haven’t established how many of the deaths listed in VAERS are caused by the vaccine, one should be skeptical of even the number 5,000. 500,000 is an impossibly large number of vaccine deaths. In another podcast, Bret and Heather suggest that the vaccine kills 2 young people for every 3 elderly people it saves. That would also imply hundreds of thousands of vaccine deaths, if it were true.
5. Discussion on ending the pandemic with ivermectin starts at 4:45.
6. The original in vitro study found the effective concentration (IC50) of ivermectin is about 2.2 μM. Doses below some threshold (maybe 1.0 μM?) should have no effect at all:
In January 2021, Dr Kory claimed that this value was lower in “alveolar cells”, only 0.41 μM, based on unpublished research:
To my knowledge, that research has still not been published. Also, Kory made multiple errors in that slide. He got the molecular weight of ivermectin wrong, it’s 875 grams per mole, not 750. And the number 105 ug/g should actually be 359 ug/g. I don’t know why he got those both wrong. Too busy saving the world, I guess?
In October 2021, a different group published that the dose required was as low as 0.2 μM in lung cells (calu-3 cell line).
That would make the drug ten times as effective as we thought it is. Then, to confuse the matter even more, a 3rd group published the opposite results in May 2021. Their experiment found that ivermectin has no measurable anti-viral effects in lung cells.
So, how do these numbers relate to dose sizes of ivermectin?
With a dose of 0.2 mg/kg (about 16 mg for an average person), taken on an empty stomach, the blood concentration peaks at 0.05 μM.
The amount that gets from your blood into your lungs is probably higher. We don’t know the ratio for humans but it’s 2.7 in cows, 3 in goats, and 1.4 in mice.
Using the largest number, the lung concentration peaks at maybe 0.15 μM. Increase the dose of ivermectin to 0.4 mg/kg, we’re at 0.3 μM. At 0.6 mg/kg, 0.45 μM. At high doses, those numbers might exceed the therapeutic concentration in lung cells.
Taking the drug with a meal should raise the concentrations higher. Some studies say the dose goes up by 150%. One study also says it only goes up by 25%. Using the larger value, treatment at 0.6 mg/kg could get you up to 1.1 μM.
There’s yet another complication, because about 93% of the drug binds to albumin in your blood. So you might need to multiply that number by 0.07, I’m not exactly sure here. Most papers seem to list both the total drug amount and the free amount. So maybe we’re back down to only 0.08 μM of free ivermectin.
One last detail is that the drug stays in your system for a long time. The half-life in blood is about 18 hours. In lung tissue, the amount appears to stay constant for 8 days and then the concentration declines. So if you take it for 5 days, the concentration in lungs should build up. You’re up to 0.16 μM on day 2 and up to 0.4 μM on day 5.
We’re trying to hit 0.2 μM in lung cells. I think that might be reachable.
But there are a few unknown numbers in this calculation — the exact dose needed to have an effect, the ratio between lung and blood doses, and the exact benefit of taking the drug with food. Pick and choose the right numbers here and it’s possible the drug hits the right therapeutic doses if you use enough of it.
Does this mean that ivermectin obviously works? No, the situation is still complicated.
In the Together trial, ivermectin showed weak results, possibly even nothing at all, using 0.4 mg/kg for 3 days. If the drug is weak in those doses, then it should also have been weak in the Mexico City trial where they only gave out 0.2 mg/kg for 2 days. But that trial claimed a 75% success rate.
The drug should also be weak when people take 0.2 mg/kg for prevention. The FLCCC used to say that you only needed to take that dose once a month to be immune. They kept changing the recommendation, now they’re all the way up to twice a week. That dose seems like it would barely work, even with the most generous assumptions.
My best take on this is we should be suspicious of any studies claiming that prevention works or that there are strong treatment results from low doses. But this math does not rule out the possibility that ivermectin works for treatment, particularly in high dose regimes (like 0.6 mg/kg for 5 days, given with a meal).
